Here's the abstract of an NBER working paper entitled Why Stars Matter by Ajay Agrawal, John McHale, Alexander Oettl:
The growing peer effects literature pays particular attention to the role of stars. We decompose the causal effect of hiring a star in terms of the productivity impact on: 1) co-located incumbents and 2) new recruits. Using longitudinal university department-level data we report that hiring a star does not increase overall incumbent productivity, although this aggregate effect hides offsetting effects on related (positive) versus unrelated (negative) colleagues. However, the primary impact comes from an increase in the average quality of subsequent recruits. This is most pronounced at mid-ranked institutions, suggesting implications for the socially optimal spatial organization of talent.
If it sounds forbiddingly academic, the Chronicle has a translation into plain English: How Hiring a Star Matters to a Department.